ObjectiveTo determine the fatality rate of horses undergoing general anaesthesia at a private equine referral practice using a limited number of anaesthetic protocols. MethodsA retrospective analysis of records (n = 17 961) from all horses undergoing general anaesthesia for surgical procedures from 1997 to 2001 at Rood and Riddle Equine Hospital, Lexington, Kentucky, USA. Results were reported as percentage of the population, and as crude mortality rates for each procedure (deaths per 1000). ResultsThe prevalence of equine fatalities directly related to anaesthesia was 0.12% (n = 21) and this rose to 0.24% (n = 42) with the inclusion of horses killed or dying within 7 days post general anaesthesia. Causes of death directly related to anaesthesia were cardiac arrest (n = 10), fracture in recovery stall (n = 8), neuropathy and myopathy necessitating euthanasia (n = 3). Crude mortality rates per procedure were ≤7 deaths per 1000 cases, except arthrodesis/osteotomy cases were 66.7 deaths per 1000 cases. ConclusionsThe anaesthetic fatality rate at this practice is lower than has been reported previously.Clinical relevance Familiarity with an anaesthetic protocol in combination with reduced anaesthetic time, emergencies of shorter duration between diagnosis and surgery, and adequate preoperative examination appear to minimize the risks associated with general anaesthesia in horses.